The current primary form of treatment for all stages of colon cancer is surgical resection, with overall survival rates depending upon whether the cancer recurs and metastasizes. The likelihood of recurrence is directly related to the stage of cancer, which is determined by the depth of tumor penetration into the wall of the colon. Recurrence rates range from 20% in stage I, up to 40% in stage II patients and up to 60% in stage III patients. However, colon cancer staging is not an accurate method for predicting recurrence. In fact, all existing approaches to postoperative surveillance of patients with colon cancer are inadequate because disease recurrence is only detected when interventions have relatively little benefit on patient survival. Several prognostic markers have been proposed but because of a lack of specificity and sensitivity these have not been approved for use as clinical markers. New colon cancer markers that could accurately predict recurrence would be valuable for identifying patients that are at risk of cancer and may benefit from adjuvant therapy shortly after resection. The long-term goal for this project is to develop a molecular diagnostic assay that can be used at the time of resection of a primary colon tumor to identify patients with a high risk of recurrence. We will focus on microRNA (miRNA) biomarkers in these studies. Within the last two decades there has been great interest in miRNAs, which are noncoding small RNAs that have been initially identified as regulators of development, and have more recently been demonstrated to have a role in a number of cancers. Additionally, miRNA expression patterns have been shown to be highly predictive of cancer type, making them ideal as prognostic indicators. Our preliminary investigations have revealed miRNAs that are highly predictive of recurrence, but these studies need to be validated and expanded. We will utilize a comprehensive miRNA expression profiling system to the expanded miRNA discovery and build biomarker sets for prediction of recurrence of CRC. Bioinformatics analyses will be used to establish prognostic indices with high performance in predicting recurrence. In phase II we will implement design features on the validated biomarker set to develop a robust, reliable, and reproducible qRT-PCR assay for miRNA analysis. We anticipate that our studies will be the foundation for a comprehensive clinical test to assist physicians and patients in the management of colon cancer recurrence. PUBLIC HEALTH RELEVANCE: Resection of colon cancer is often not effective because patients with significantly advanced disease have a high likelihood of recurrence and metastasis. Our long term goal is to develop a prognostic assay for colon cancer recurrence that will improve clinical management of colon cancer by identifying patients prior to clinical manifestation of recurrence, and at a stage at which medical intervention may benefit the patient.